Susan C. Miller, PhD

Hospice Care in Nursing Homes and Policy Issues
Susan C. Miller, PhD

A half million older adults die in U.S. nursing homes each year. Hospice care, certified and reimbursed by Medicare, offers specialized end-of-life care in the nursing home when nursing homes contract and collaborate with hospice providers. There has been much growth in the use of nursing home hospice care (and in average lengths of stay), which has been accompanied by greater costs to the Medicare hospice program. These greater hospice costs (in and outside of nursing homes), and a 50 percent growth in the number of (mostly for-profit) Medicare-certified hospice providers, raised policy concerns which lead to review by the Medicare Payment Advisory Commission, or MedPAC.

MedPAC commissioned analyses which resulted in a recommendation that there be a strengthening of procedures for determining recertification of hospice eligibility. (It recommended procedures for greater provider accountability and more Office of Inspector General investigations). MedPAC also recommended a changed Medicare payment system that would vary per diem payments to better reflect the differing intensity of hospice service provided across hospice episodes of care. (MedPAC, 2009) Under the Patient Protection and Affordable Care Act of 2010, Congress gave CMS the authority to revise, in a budget-neutral manner, the payment system, but it did not mandate it be changed in the manner recommended by MedPAC. CMS has just released an RFP to study creation of revised payment system.

To inform efforts in Medicare hospice reform, we undertook a number of studies at the Center for Gerontology and Health Care Research at Brown University. Specifically, given our ability to expertly link nursing home resident assessment (MDS) data with Medicare enrollment and claims data (Intrator et al., 2011), we aimed to conduct analyses of nursing home-linked data to inform policy regarding hospice care in nursing homes. Below I highlight some of our research findings.

Previous research has found the use of hospice care to augment nursing home end-of-life care to be beneficial.(Gozalo & Miller, 2007; Miller et al., 2002) However, its use has been historically low. However, while in the 1990s a low proportion of U.S. nursing homes contracted with a hospice, recent analyses show this proportion to now be closer to 90 percent. We have also found a doubling of hospice use by nursing home residents between 1990 and 2006, and in 2006 one-third of Medicare beneficiaries who died in nursing homes accessed hospice. (Miller et al, 2010). For nursing home residents who die with dementia, use of Medicare hospice care nearly tripled, with approximately 40 percent of decedents accessing hospice in 2006. (Miller et al., 2011) Additionally, of the nursing home residents who died in 2006 and received hospice, only 17 percent had cancer diagnoses.

While the increased use of hospice care in nursing homes is viewed as a positive development, this increased use has been accompanied by longer average lengths of stays and thus greater costs to the Medicare hospice program; and, average lengths of stay are longer for nursing home versus non nursing home hospice patients. (Miller et al., 2010) In fact, for residents dying with advanced dementia in nursing homes, average lengths of stay have tripled -- from 46 days in 1999 to 118 days in 2006. Averages, though, are misleading, and what we are really seeing is a growth (in some states) in the proportion of long hospice stays (>180 days), and nationally an increase in the length of long hospice stays (i.e., stays at the 90th percentile of the distribution are longer). Notable, however, is that hospice length of stay distributions are bimodal, and approximately 25 percent of hospice patients continue to have stays of seven days or less.

In addition to the above, we have observed significant differences across U.S. states in the use of nursing home hospice. In conjunction with this, the volume of growth in hospice use within a state is associated its growth of Medicare certified hospice providers. Furthermore, hospice average lengths of stay are longer in states with the greatest provider growth (versus those with lower growth).

Based on the above and other analyses, we concluded that there is support for MedPAC’s recommendation to modify the current Medicare hospice reimbursement system. Also, It appears that varying payments as a function of length of stay may be a logical approach for reducing lengths of stay (some of which are believed to be provider driven) and Medicare costs. However, we cautioned (in our published work) that it is important for any new policy to explicitly acknowledge the challenges inherent in the timing of hospice referral for nursing home residents in the last stages of (noncancer) chronic terminal illnesses by recognizing “early” referrals leading to long hospice stays will occur. We believe these longer stays should be deemed “acceptable” in the presence of well-documented physician evaluations and eligibility determinations. Without such explicit acknowledgement, the fear is that undue scrutiny may occur, resulting in decreasing enrollments and a higher prevalence of very short hospice stays.

The above work was supported by grants from the Agency for Health Care Research and Quality (R03HS016918), the Shaping Long Term Care in America Project funded by the National Institute on Aging (P01AG027296), and by an Alzheimer’s Association grant (IIRG-08-91343). Our continuing research is examining the effect of nursing home hospice use (at the resident and nursing home level) on health care utilization and quality outcomes. Also, Dr. Gozalo and colleagues have a contract with the National Hospice Foundation of the National Palliative Care Organization to examine changes to the Medicare Hospice Benefit Payment System, using visit data from for- and non-profit hospices. Last, through our NIA-funded Shaping Long Term Care in America Project, we have created a website that has much hospice and nonhospice nursing home-, county- and state-level data (go to

Gozalo P, Miller S. Hospice enrollment and evaluation of its causal effect on hospitalization of dying nursing home patients. Health Serv Res. 2007;42: 587-610.

Intrator O, Hiris J, Berg K, Miller SC, Mor V. The Residential History File: Studying Nursing Home Resident’s Long Term Care Histories, Health Services Research, 2011;46: 120-137.

Medicare Payment Advisory Commission. Reforming Medicare's Hospice Benefit. Report to the Congress: Medicare Payment Policy. Washington, DC, 2009, pp. 347-376.

Miller SC, Lima J, Gozalo P, Mor V. The Growth of Hospice Care in U.S. Nursing Homes, Journal of American Geriatrics Society, 2010;58:1481-88.

Miller SC, Lima J, Mitchell SL. Hospice Care for Persons with Dementia: The Growth of Access in U.S. Nursing Homes. Journal of Alzheimer’s Disease and Other Dementias, 2010;25: 666-73.

Miller SC, Mor V, Wu N, Gozalo P, Lapane K. Does receipt of hospice care in nursing homes improve the management of pain at the end of life? J Am Geriatr Soc. 2002;50: 507-515.